The explosive growth of Internet and the recent increasing trends in automation using intelligent appli- cations have provided a veritable playground for malicious software (malware) attackers. With a variety of devices connected seamlessly via the Internet and large amounts of data collected, the escalating mal- ware attacks and security risks are a big concern. While a number of malware detection methods are available, new methods are required to match with the scale and complexity of such a data-intensive environment. We propose a novel and unified hybrid deep learning and visualization approach for an effective detection of malware. The aim of the paper is two-fold: 1. to present the use of image-based techniques for detecting suspicious behavior of systems, and 2. to propose and investigate the application of hybrid image-based approaches with deep learning architectures for an effective malware classification. The performance is measured by employing various similarity measures of malware behavior patterns as well as cost-sensitive deep learning architectures. The scalability is benchmarked by testing our proposed hybrid approach with both public and privately collected large malware datasets that show high accuracy of our malware classifiers.
Keywords: Malware detection | Similarity mining | Image analysis | Evaluation metrics | Machine learning | Deep learning architectures

The distance and similarity measures are two interrelated depictions of the patterns which signify the categorization between the Atanassov intuitionistic fuzzy sets (AIFSs) by evaluating the degree of belongingness. In this work, we propose a new similarity measure, termed as hybrid similarity measure, by the combination of intuitionistic fuzzy bounded variation (IFBV) and intuitionistic fuzzy metric based measures. The concept of IFBV which is a technique to approximate arc length of an intuitionistic fuzzy-valued function (IFVF) is also introduced here. The IFVF over an AIFS is geometrically evolved through the generalization of the p-summable IFBV, that is, connecting all the elements of AIFS lie on the structure of IFBV corresponding to power p. The proposed measure overcomes the shortcomings of intuitionistic fuzzy metric based similarity measures by incorporating more flexibility into it. The hybrid similarity measure has been successfully implemented and applied on the several real-world applications in the field of pattern recognition as well as clustering. Further, a detailed comparison of results has been shown against the other existing similarity measures to demonstrate the superiority and validity of the proposed hybrid similarity measure.
Keywords: Intuitionistic fuzzy set (IFS) | Similarity measure | Bounded variation | Pattern recognition | Clustering analysis

In general, a practical online recommendation system does not rely on only one algorithm but adopts dif- ferent types of algorithms to predict user preferences. Although most of similarity measures can rapidly calculate the similarity on the basis of co-rated items, their prediction accuracy is not satisfactory in the case of sparse datasets. Making full use of all the rating information can effectively im prove the rec- ommendation quality, but it reduces the system efficiency because all the ratings need to be calculated. To recommend items for target users rapidly and accurately, this paper designs a hybrid item similarity model that achieves a trade-offbetween prediction accuracy and efficiency by combining the advantages of the two above-mentioned methods. First, we introduce an adjusted Google similarity to rapidly and precisely calculate the item similarity in the condition of enough co-rated items. Subsequently, an intu- itionistic fuzzy set (IFS) based Kullback–Leibler (KL) similarity is presented from the perspective of user preference probability to effectively compute the item similarity in the condition of rare co-rated items. Finally, the two proposed schemes are integrated by an adjusted variable to comprehensively evaluate the similarity values when the number of co-rated items lies in a certain range of value. The proposed model is implemented and tested on some benchmark datasets with different thresholds of co-rated items. The experimental results indication that the proposed system has a favorable efficiency and guarantees the quality of recommendations
Keywords: Recommender system | Collaborative filtering | Normalized Google distance | Intuitionistic fuzzy set | Kullback–Leibler divergence

Air traffic managers face challenging decisions due to uncertainity in weather and air traffic. One way to support their decisions is to identify similar historical days, the traffic management actions taken on those days, and the resulting outcomes. We develop similarity measures based on quarter-hourly ca pacity and demand data at four case study airportsdEWR, SFO, ORD and JFK. We find that dimensionality reduction is feasible for capacity data, and base similarity on principal components. Dimensionality reduction cannot be efficiently performed on demand data, consequently similarity is based on original data. We find that both capacity and demand data lack natural clusters and propose a continuous similarity measure. Finally, we estimate overall capacity and demand similarities, which are visualized using Metric Multidimensional Scaling plots. We observe that most days with air traffic management activity are similar to certain other days, validating the potential of this approach for decision support.
Keywords: Similar days | Clustering capacity and demand data | Decision support | Air traffic management

Evaluating similarity between trajectories of moving objects is important for wide range of applications. The existing similarity measures typically define some meaning of similarity and propose algorithms for computing it. We think that the meaning of similarity is application dependant, and should only be determined by the user. Therefore, there is a need for a generic approach where users can define the meaning of similarity. In this paper, we propose a parametrized similarity operator, based on the time warped edit distance, where the meaning of similarity is generic and left for user to define. Our proposed operator is implemented in SECONDO and evaluated using both synthetic and real datasets. The results were promising and as expected.
KEYWORDS : Trajectory similarity | Moving objects databases | SECONDO | TWED